Hospital Mortality Prediction

Predicting in-hospital mortality aims to identify patients at high risk of death, enabling timely interventions and improved resource allocation. Current research emphasizes developing more accurate and explainable prediction models, leveraging diverse data sources like electronic health records (EHRs), clinical notes, and physiological time series, often employing advanced architectures such as transformers, neural ODEs, and multimodal learning approaches. These efforts focus on addressing data irregularities, handling missing values, and improving model interpretability to build trust and facilitate clinical adoption. Ultimately, improved mortality prediction holds significant potential to enhance patient care and optimize healthcare resource management.

Papers

July 12, 2024

Towards Personalised Patient Risk Prediction Using Temporal Hospital Data Trajectories
Thea Barnes, Enrico Werner, Jeffrey N. Clark, Raul Santos-Rodriguez
Risk Prediction Early Warning Patient Trajectory Hospital Mortality Prediction

July 6, 2024

BadCLM: Backdoor Attack in Clinical Language Models for Electronic Health Records
Weimin Lyu, Zexin Bi, Fusheng Wang, Chao Chen
Backdoor Attack Electronic Health Record Clinical Decision Support Hospital Mortality Prediction Clinical Language Model

February 9, 2024

A Kalman Filter Based Framework for Monitoring the Performance of In-Hospital Mortality Prediction Models Over Time
Jiacheng Liu, Lisa Kirkland, Jaideep Srivastava
System Performance Time Matter Kalman Filter Class Distribution Sample Size Precision Recall Curve Hospital Mortality Prediction

December 29, 2023

XAI for In-hospital Mortality Prediction via Multimodal ICU Data
Xingqiao Li, Jindong Gu, Zhiyong Wang, Yancheng Yuan, Bo Du, Fengxiang He
Multimodal Phenomenon Multimodal Dataset xAI Community Multimodal Input Intensive Care Unit Hospital Mortality Prediction

December 11, 2023

Multimodal Pretraining of Medical Time Series and Notes
Ryan King, Tianbao Yang, Bobak Mortazavi
Self Supervised Learning Self Supervised Pretraining Short Note Multimodal Pre Intensive Care Unit Medical Time Series Hospital Mortality Prediction

August 19, 2023

Contrastive Learning-based Imputation-Prediction Networks for In-hospital Mortality Risk Modeling using EHRs
Yuxi Liu, Zhenhao Zhang, Shaowen Qin, Flora D. Salim, Antonio Jimeno Yepes
Imputation Algorithm Patient Representation Deep Imputation EHR Datasets Hospital Mortality Prediction

December 12, 2022

Mortality Prediction Models with Clinical Notes Using Sparse Attention at the Word and Sentence Levels
Miguel Rios, Ameen Abu-Hanna
Self Attention Attention Mechanism Sentence Level Clinical Note Sparse Attention Real Text Word Clinical Prediction Model Hospital Mortality Prediction Mortality Prediction Model

August 9, 2022

A Multimodal Transformer: Fusing Clinical Notes with Structured EHR Data for Interpretable In-Hospital Mortality Prediction
Weimin Lyu, Xinyu Dong, Rachel Wong, Songzhu Zheng, Kayley Abell-Hart, Fusheng Wang, Chao Chen
Clinical Note Multimodal Transformer Clinical Narrative Hospital Mortality Prediction Clinical BERT

August 5, 2022

COPER: Continuous Patient State Perceiver
Vinod Kumar Chauhan, Anshul Thakur, Odhran O'Donoghue, David A. Clifton
Perception Model Patient Representation Patient Trajectory Continuous Time Dynamic Hospital Mortality Prediction

Hospital Mortality Prediction

Papers

Towards Personalised Patient Risk Prediction Using Temporal Hospital Data Trajectories

BadCLM: Backdoor Attack in Clinical Language Models for Electronic Health Records

A Kalman Filter Based Framework for Monitoring the Performance of In-Hospital Mortality Prediction Models Over Time

XAI for In-hospital Mortality Prediction via Multimodal ICU Data

Multimodal Pretraining of Medical Time Series and Notes

Contrastive Learning-based Imputation-Prediction Networks for In-hospital Mortality Risk Modeling using EHRs

Mortality Prediction Models with Clinical Notes Using Sparse Attention at the Word and Sentence Levels

A Multimodal Transformer: Fusing Clinical Notes with Structured EHR Data for Interpretable In-Hospital Mortality Prediction

COPER: Continuous Patient State Perceiver